Dynamic Analysis of Numerical Mesoscale Models of Composite Materials

• Autor(a) principal: Silva, André Alves
• Data de Publicação: 2016
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/19717
• topic: Delamination; Hourglass; Failure criterion; Explicit dynamic analysis; Mesoscale; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Mecânica

The role of composites in the industry has been increasing over the past few years. One of the main reasons for this phenomenon is due to these materials presenting the best properties of their constituents and often qualities that neither of the constituent materials have. However, as a consequence, composite structures are becoming more complex, and its behavior more difficult to predict. The study of this dissertation arises from Crespo’s dissertation, and had its origin on the need to predict the behavior of composite materials specimens in a less expensive and faster manner. To fulfill this need, a methodology was developed using LS-DYNA finite element mesoscale models. This methodology uses an explicit dynamic analysis, and is capable of testing composite specimen models subjected to tensile and compressive loads. However, this type of analysis is different from the one Crespo used and phenomenons like hourglass must be considered. For this case study the material model consists in a spread tow carbon fabric with 0°/90°, 15°/75° and 30°/60° arrangements. To simulate the behavior of this composite material, when subjected to external loads, failure and damage propagation such as delamination, fiber and matrix failure are implemented in the models. The failure of the matrix and the fibers are controlled by a combined failure criterion implemented in LS-DYNA code. Knowing only the dimensions of the test specimens, the material properties were taken from Crespo’s dissertation. To validate the models, results were compared with Mangualde’s, whose methodology was developed using an implicit static analysis.